IV. REACTIVITY AND SLEEP IN INFANTS:
A LONGITUDINAL OBJECTIVE ASSESSMENT
Gali S. De Marcas, Nirit So¡er-Dudek, Shaul Dollberg,Yair Bar-Haim,
and Avi Sadeh
ABSTRACT Sleep patterns and temperament in the first year of life are
closely related. However, research utilizing objective, rather than subjective
measurements of sleep and temperament is scarce and results are inconsistent.
In addition, a relative lack of longitudinal data prevents inference of causality
between the two constructs. In this study, infant sleep was objectively assessed
among 95 infants at 3, 6, and 12 months-of-age with an actigraph in the home
setting. Reactivity to sound, light, and touch, a specific aspect of temperament, was behaviorally assessed at 3 and 6 months, both during sleep (at
home) and during waking (at the laboratory). Expected maturational trends
were recorded in sleep, with a temporal increase in sleep efficiency and
percent of motionless sleep. Quadratic (i.e., inverse U shape) relations were
found, especially among girls, when predicting change in sleep by reactivity
thresholds, suggesting that both hyposensitive and hypersensitive infants are
at risk for poor sleep quality. These are the first research findings suggesting
that low reactivity in infancy might be associated with compromised sleep
quality. The observed nonlinear effects may account for null or inconsistent
results in previous studies that explored only linear associations between
temperament and sleep. Future studies should address both extremes of the
temperament continuum when exploring relations with sleep patterns.
Sleep and temperament in infancy are closely linked and both have
important developmental implications (Ednick et al., 2009). Within the
systems perspective, both sleep and temperament are among the child’s
intrinsic characteristics that exert influence on and are affected by the child
context (see El-Sheikh & Sadeh, Figure 1, Chapter I, in this volume).
Early sleep problems are related to psychopathology and behavioral
difficulties later in development (see Ednick et al., 2009; Gregory & Sadeh,
Corresponding author: Avi Sadeh, School of Psychological Sciences, Tel Aviv University,
Tel Aviv 69978, Israel, email: [email protected]
This research was supported by the Israel Foundation Trustees.
49
Baseline sleep
(3M or 6M)
Predicted sleep
(6M or 12M)
Reacvity (linear)
(3M or 6M)
Reacvity (curvilinear)
(3M or 6M)
FIGURE 1.—The models tested. Three prospective models were tested: from age
3 months to 6 months, from age 3 months to 12 months, and from age 6 months to 12 months.
2012, for reviews). Sleep disorders in young children are persistent; for
instance, 41% of children who had sleep difficulties at 8 months-of-age still
had them at age 3 years (Zuckerman, Stevenson, & Bailey, 1987). Early
screening of sleep-related issues may be useful for prevention of cognitive and
behavioral deficits (Ednick et al., 2009). In addition, sleep problems may
cause fatigue in the parents and disrupt family functioning (Sadeh, Mindell,
& Owens, 2011; see Tikotzky et al., Chapter VII, in this volume).
Temperament is another key construct in early child development.
Difficult temperament has been described as a characteristic of infants who
tend to be very emotional, irritable, and fussy, and cry a lot and is found in
about 10% of the infants (Thomas & Chess, 1977). Infants with difficult
temperaments have been reported to receive less optimal caregiving,
rendering them at risk for behavioral problems and other adverse
developmental outcomes (Allen & Prior, 1995; Gjone & Stevenson, 1997;
Thomas, Chess, & Korn, 1982). In addition, temperament has been linked to
various psychopathologies such as Attention Deficit Hyperactivity Disorder
(Martel & Nigg, 2006), affective disorders (Mezulis, Hyde, & Abramson,
2006), anxiety symptoms (Kagan, Snidman, Zentner, & Peterson, 1999), and
attachment issues (Zeanah & Fox, 2004).
The construct of temperament includes a number of different infant
characteristics including activity level, approach-withdrawal tendencies,
mood, adaptability, and sensory thresholds. In the present study, we focus
on a key component of temperament—reactivity to external stimulation,
reflecting infants’ sensory threshold or the stimulus intensity level needed to
evoke a discernible response (Chess & Thomas, 1996). It has been argued that
the concept of temperament is largely based on two processes: reactivity to
sensory stimuli and the ability to regulate such reactivity (Rothbart &
50
REACTIVITY & INFANT SLEEP
Derryberry, 1981). Reactivity to sensory stimuli includes motor, affective,
and physiological reactions. Infants differ in their response thresholds
(Korner, 1973), and these thresholds have been related to cognitive ability
(Derryberry & Rothbart, 1997) and to psychopathology and personality
(Kagan & Snidman, 1991; Kagan et al., 1999) later in childhood. Thus,
reactivity thresholds have been considered a predisposition for personality
development during childhood (Fox & Calkins, 1993).
Temperament is related to infant sleep patterns and sleep quality (Ednick
et al., 2009). Several studies have shown that infants and toddlers with
reported sleep problems were more likely to be classified as more difficult in
temperament than other children (Atkinson, Vetere, & Grayson, 1995;
Jimmerson, 1991; Kelmanson, 2004; Shaefer, 1990; Weissbluth, 1981). With
a few exceptions, most findings in this field rely on subjective parental
reports for both sleep and temperament. However, the validity of parental
reports on infant sleep has been criticized on various grounds related to
parents’ potential reporting biases and their limited knowledge and
awareness (Sadeh, 1994, 1996; Scher, Epstein, Sadeh, Tirosh, & Lavie,
1992; see also Sadeh, Chapter III, in this volume). Similarly, parental reports
of infant temperament have been criticized because of their subjective nature
and vulnerability to perceptual and reporting biases (Keener, Zeanah, &
Anders, 1988).
In a review of 10 studies on sleep and temperament in the first year of
life (Ednick et al., 2009), sleep was objectively assessed in only one-half of
the studies, and temperament was assessed with subjective self-reports in all
10 studies. Inconsistent results emerge from studies that used objective sleep
assessments along with subjective temperament assessments. For example, a
longitudinal study with objective sleep assessment and subjective reports of
infant temperament showed that increased sleep in the first year of life was
related to an “easy” temperament (Spruyt et al., 2008). In contrast, Scher,
Tirosh, and Lavie (1998) concluded that their results do not support a link
between sleep regulation and quality and temperament in infancy.
Studies on infant temperament and sleep using objective measures are
scarce, and their results are inconsistent. One longitudinal study objectively
measured sleep and temperament and used a parent-reported temperament
questionnaire (Halpern, Anders, Garcia Coll, & Hua, 1994). Sleep patterns at
3 weeks-of-age and at 3 months-of-age were related to behavioral temperament assessments at 3 months-of-age; for example, more time awake during
the night at 3 weeks-of-age was related to both higher observed irritability
scores and higher observed inhibition scores at 3 months-of-age. In this study,
maternal reports of temperament were mostly unrelated to sleep indices
(Halpern et al., 1994).
As indicated above, the focus of our study was on reactivity or sensory
threshold; a key component of temperament. Carey (1974) suggested that the
51
relation between temperament and sleep might stem from an underlying
physiological reactivity factor. Such a factor could cause both difficult
temperament and poor sleep. That is, a lower sensory threshold could lead to
difficulties in regulating stimulation during wakefulness (difficult temperament) and also difficulties in disengaging from external and internal
stimulation needed for initiating and maintaining sleep (Carey, 1974). Sadeh
et al. (1994) suggested two additional explanations for the link between
temperament and sleep in young children. First, sleep fragmentation might
impair emotional and cognitive regulation, causing hypervigilance, distractibility, and other characteristics of difficult temperament. Second, parental
styles in domains such as limit setting might act as an external factor bringing
about both sleep problems and behaviors characteristic of difficult
temperament. In this study, we focused on Carey’s physiological threshold
explanation, that is, we explored the possibility that sensory reactivity predicts
subsequent changes in sleep quality. We measured both sensory reactivity
and sleep using objective assessment methods. Sleep was assessed at 3, 6, and
12 months of age and sensory reactivity was tested at 3 and 6 months of age.
These ages were chosen as good representative points of the process of sleep
consolidation and reactivity changes (e.g., Coll, Halpern, & Vohr, 1992;
Henderson, France, Owens, & Blampied, 2010).
To our knowledge, sensory reactivity during sleep and its relations to
regular sleep quality has not been assessed in infants. Most studies assessing
reactivity during sleep focused on an auditory arousal threshold (e.g., Busby,
Mercier, & Pivik, 1994; Franco et al., 1998). Furthermore, differences between
boys and girls at this age in temperament (Mezulis et al., 2006), response
threshold (Klein, 1982), and sleep (Adams, Jones, Esmail, & Mitchell, 2004;
Scher & Cohen, 2005), and in patterns of relations among activity, arousal,
and sleep (Fisher & Rinehart, 1990) suggest that that gender may play a
significant role that should be explored.
The literature on temperament emphasizes the role of both overregulation and underregulation in temperament (Derryberry & Rothbart,
1997). Derryberry and Rothbart suggest that both a lack of inhibition
(impulsivity) and overinhibition (fearfulness) may lead to social and
emotional difficulties. In other words, both extremes of the temperament
continuum may lead to adjustment problems. The same principle may apply
to links between sensory reactivity and sleep. We did not find earlier research
on sleeping patterns in hyposensitive or overly inhibited children. However,
Minard and colleagues found a quadratic association between cyclicity of
sleep at 6 months and infant mental performance scores at 12 months
(Minard, Freudigman, & Thoman, 1999). In addition, Boismier, Chappell,
and Meier (1974) noted a quadratic relation between waking activity level and
REM sleep in neonates. This literature underscores the need to explore
nonlinear links between sleep and temperament.
52
REACTIVITY & INFANT SLEEP
Aims and Hypotheses
The purpose of the current study was to conduct a thorough examination
of the links between sleep and sensory reactivity. Our specific aims were: (a) to
examine to links of sleep and sensory reactivity longitudinally during the first
year; (b) to use objective measures of both sleep and sensory reactivity; (c) to
assess sensory reactivity during sleep as well as during wakefulness; (d) to
explore the role of gender in moderating the links between sleep and sensory
reactivity; and (e) to examine these links using both linear and nonlinear
methods. Our main hypotheses were: (a) very high, and perhaps very low,
sensory reactivity would predict lower sleep quality; (b) both reactivity during
wakefulness and during sleep would be linked to sleep quality; and (c) gender
would play a role in moderating these links.
Our main hypotheses were: (a) high, and possibly very low, sensory
reactivity would predict lower sleep quality; (b) sensory reactivity during sleep
as well as during wakefulness would be linked to sleep quality; and (c) gender
would play a role in moderating the links between sleep and sensory reactivity.
METHOD
Participants
Participants were 95 full-term infants (43 female). Mean gestational age
was 39.55 weeks (SD ¼ 1.26), all had a second Apgar score of 9 or 10, and mean
birth weight was 3.21 kg (SD ¼ 0.46). None of the infants had chronic or acute
medical problems. Table 1 describes sample characteristics. The socioeconomic characteristics (education level of parents, number of children in the
family, and number of rooms in the house) suggest that the sample was mostly
representative of middle and upper-middle socioeconomic status.
Ten participants did not have laboratory temperament scores at age
3 months, and 13 did not have those scores at age 6 months. One participant
did not have a home temperament score at age 3 months, and two participants
did not have that score at age 6 months. Five, two, and ten participants did not
have sleep scores at age 3, 6, and 12 months, respectively. Missing data
resulted from technical failures or inability of infants (or parents) to complete
the procedures within the predetermined infant’s age window. We did not
impute missing data, all available data were used for each analysis.
Procedures
The study was approved by the university Ethics Committee and by the
hospital Helsinki Committee. Participants were recruited at a hospital
maternity ward. Informed consent was obtained from mothers following
53
TABLE 1
SAMPLE CHARACTERISTICS
Variable
Mean SD
Range
Gestational age (weeks)
Birth weight (kg)
Head circumference (cm)
Mother’s age (years)
Father’s age (years)
Mother’s education (years)
Father’s education (years)
Number of children in family
Number of rooms in the house
39.66 1.30
3.22 .46
34.53 2.49
32.50 4.35
34.89 5.59
16.17 2.40
15.71 2.56
1.94 .92
3.64 .90
37–42
2.11–4.35
31–53
24–42
23–61
12–25
11–30
1–4
2–7
description of the study and its aims. Infants were assessed three times: at ages
3 months (Wave 1; W1), 6 months (Wave 2; W2), and 12 months (Wave 3; W3).
W1 and W2 assessments were conducted both at participants’ homes and at a
laboratory, and W3 assessments were conducted at participants’ homes. Sleep
was measured at home in all the assessment waves, whereas sensory reactivity
was measured at W1 and W2, both at the laboratory (waking reactivity) and at
home (sleep reactivity). For each assessment wave, parents were instructed to
attach the actigraph to their child’s left ankle for at least four nights and to
keep a sleep diary. On the fifth night of each assessment wave, a home visit was
conducted in which the actigraph was collected. At W1 and W2, reactivity
during sleep was assessed.
Measures
Sleep Assessments
Activity monitoring with complementing daily sleep diaries were used to
assess sleep–wake patterns. Sleep data were collected for four consecutive
nights using actigraphs (AMA-32, Ambulatory Monitoring Inc., Ardsley, NY).
Actigraphy has been established as a reliable and valid method for naturalistic
study of sleep in infants, children, and adults (Sadeh, 1994; Sadeh, Acebo,
Seifer, Aytur, & Carskadon, 1995; Sadeh, Hauri, Kripke, & Lavie, 1995). At age 3
months (W1), the actigraph was attached for 24 hours, but sleep parameters
were calculated only for nights between 8 p.m. and 8 a.m. These boundaries,
which are somewhat arbitrary, were based on reviewing the data from this
study’s sample as well as earlier samples. These boundaries to the definition
of nocturnal sleep have been established in previous research to create
standards that can be compared across studies (Sadeh, 2004). At 6 and 12
months (W2, W3), the actigraph was attached at night only. Our past
experience has led us to realize that having infants at this age continuously wear
54
REACTIVITY & INFANT SLEEP
actigraphs, during a time when infants start crawling and become more aware
of their body, leads to many failures in compliance. Furthermore, daytime naps
are often occurring in unpredictable places and in circumstances that lead to
actigraphy artifacts (e.g., baby sleeping in a moving vehicle or stroller or in his
parent’s arms, creating externally induced movement artifacts). Therefore we
decided to focus on the main nocturnal sleep period that has been the focus
of previous research as well. Sleep variables were based on actigraphic data
using a validated algorithm (Sadeh, Lavie, Scher, Tirosh, & Epstein, 1991). The
following parameters were used for this study: (a) sleep efficiency, defined as
time spent asleep out of the total sleep period; and (b) motionless sleep—
percent of quiet sleep time with no activity. These measures were chosen as they
represent measures of sleep quality that are comparable from 3 to 12 months of
age. We did not include measures of sleep duration as they were not linked to
temperament in any of past studies on the links between reactivity and sleep.
Sleep diaries were used to identify and remove artifacts and to verify schedules
before sleep–walk algorithms were applied to the actigraphy data (Acebo et al.,
2005; Acebo et al., 1999).
Sensory Reactivity Assessments
Sensory reactivity levels were assessed during sleep and wakefulness.
Assessment of sleep reactivity took place at the infant’s home during a night
visit. Once infants were in their first quiet sleep episode after nocturnal sleep
onset they were stimulated with tactile, auditory, and light stimuli. Tactile
stimulation consisted of 10 hairbrush strokes on the infant’s temple, each
lasting 1 second. Sound stimuli consisted of human voices recorded at a
playground. The sounds were delivered through speakers placed one meter
away from the infant’s ears. Sound was presented at three intensities (70, 80,
and 90 dB hearing level). Rise time of the stimulus was about 2 seconds. Sound
intensity was monitored using a digital sound meter (RadioShack, Model:
33-2055). Light stimuli were presented with three intensities: 10-15 Lux, 340345 Lux, 595-600 Lux. Light was produced by a halogen lamp (Zf-L300/500p,
230v-50Hz, Max500w, IP54) connected to a dimmer and a white plastic
lampshade. The lamp was placed one meter away from the infant’s head.
Light intensity was monitored using Digital Lux Tester (YF-1065, Test Lab).
Light presentations lasted 10 seconds each. At 3- and 6-months of age, 15 and
18 infants, respectively, woke up at some point during the test.
Infant’s reactions to all stimuli were coded for an additional 10 seconds. A
10-second interval elapsed before presenting the next stimulus. The test was
stopped if the infant awoke. Sequences of stimuli (tactile, sound, and light)
were counterbalanced. Infants were examined during quiet sleep (no eye or
body movement for 5 minutes before the start of the procedure). Trained
observers (blind to all other parameters of the study) recorded infant’s
responses (e.g., eye opening, startle, crying, arm and leg movements, sucking,
55
head movement, change in body position). Inter-rater reliability, based on
parallel live scoring by two observers, was acceptable (kappa ¼ .80, p < .001).
The number of discrete responses (e.g., cry, movement, eye opening) created
a reactivity score for each stimulus type (tactile, sound, light). Based on the
methods described by Kisilevsky and Muir (1984), the scoring was based on
frequency of responses and each response received a score of 1 and then
responses were summed across situations. The reactivity scores were averaged
to create an overall reactivity score during sleep (REACT-SLP).
Reactivity during the waking state was assessed in the laboratory. Testing
was conducted in a room (3.5 m 4 m). The infant was initially placed
unstrapped in an infant car safety seat. As described above, tactile, auditory,
and light stimuli were presented and the infant’s reactivity was recorded with
one difference: presentation of light was initiated with a low amount of light
(5 Lux) rather than in darkness to avoid frightening the infants. These
reactivity scores were averaged to create an overall reactivity score during
wakefulness (REACT-WK). Inter-rater reliability for the awake condition was
acceptable (kappa ¼ .82, p < .001). Additional subtypes of waking responses
that included coding gross motor movements such as arm waving, kicking, and
back arching were also scored (REACT-MTR).
If an infant cried for more than 10 seconds, a researcher would attempt
to soothe him or her gradually, using the following sequence: (1) presenting
the infant with a toy for 10 seconds; (2) talking to the infant for 10 seconds;
and (3) caressing the infant for 10 seconds. Once the infant was soothed
testing proceeded as planned. If the infant persisted in crying, the testing
was stopped and parents were invited into the room to sooth their child. At
3 months of age, three infants needed soothing to resume the test and nine
could not be soothed back to testing. At 6 months of age, one infant required
soothing to continue and 13 needed to stop the testing.
Plan of Analysis
Univariate analyses for sleep parameters and reactivity were conducted.
We also screened reactivity and sleep variables for univariate outliers.
Individual scores higher than Z ¼ 3.29 (corresponding to a probability of
99.9%), separate from the rest of the distribution, were considered extreme
outliers that might cause distortion in regression analyses (Tabachnick &
Fidell, 2007). There were a total of six extreme scores. These were distributed
among periods and measures with no more than one score per measure, per
period needing correction. As suggested by Tabachnick and Fidell (2007), the
Winsorising procedure was used and these raw scores were changed to equal
one unit above the next highest score within that distribution.
To predict changes in sleep by sensory reactivity scores, we conducted
regressions of data collected from age 3 months to 6 months, 3 months to
56
REACTIVITY & INFANT SLEEP
12 months, and 6 months to 12 months. As mentioned earlier, we opted to
analyze data for boy and girl infants separately, to identify associations
between reactivity and sleep that might vary by gender. Thus, all predictors
were centered separately for boys and for girls. Each regression equation
predicted a sleep outcome (sleep efficiency or motionless sleep), with the
following predictors: baseline (earlier time point) sleep (sleep efficiency or
motionless sleep), a reactivity variable (REACT-SLP, REACT-WK, or REACTMTR), and the squared score of the respective reactivity variable (to
investigate nonlinear effects). Controlling for baseline sleep scores in the
regression analyses means that what is predicted is the change over time in
sleep patterns. The Apgar score and gestational age did not contribute to the
explained variance and were removed from the primary analyses. Figure 1
illustrates the model tested. The term baseline in this model (and in Figure 1)
relates to the earlier time point (e.g., 3-month measures serve as baseline for
6-month measures).
RESULTS
Sleep Univariate Analyses
Figure 2 depicts means and standard errors of sleep efficiency (2a) and
motionless sleep (2b) for boys and girls at all three assessment waves. Stability
correlations for sleep efficiency were r ¼ .10, ns; r ¼ .25, p ¼ .02; and r ¼ .01,
ns for W1-W2, W2-W3, and W1-W3, respectively. Stability correlations for
motionless sleep were r ¼ .40, p < .001; r ¼ .38, p < .001; and r ¼ .23, p ¼ .04 for
W1-W2, W2-W3, and W1-W3, respectively. Repeated measures ANOVAs,
controlling for between-subjects gender differences, revealed that there were
significant developmental changes in both sleep efficiency and motionless
sleep across the three measurement waves. For sleep efficiency, the repeated
measures effect of time was statistically significant (F[2,156] ¼ 178.55, p < .001,
partial h2 ¼ .70), and tests of within-subject contrasts revealed that sleep
efficiency at W2 was higher than at W1 (F[1,78] ¼ 169.95, p < .001, partial
h2 ¼ .69), and that at W3 was higher than at W2 (F[1,78] ¼ 11.48, p ¼ .001,
partial h2 ¼ .13). The effect of gender was also statistically significant
(F[1,78] ¼ 5.52, p ¼ .02, partial h2 ¼ .07) with girls exhibiting higher sleep
efficiency than boys. The two-way interaction between gender and time was
not significant.
For motionless sleep, the repeated measures effect of time was also
statistically significant (F[2,156] ¼ 22.16, p < .001, partial h2 ¼ .22), and tests of
within-subject contrasts revealed that at W2 motionless sleep was higher than
at W1 (F[1,78] ¼ 45.85, p < .001, partial h2 ¼ .37), but at W3 it was lower than at
W2 (F[1,78] ¼ 11.49, p ¼ .001, partial h2 ¼ .13). Gender and the two-way
57
FIGURE 2.—Means and standard errors of sleep efficiency (a) and motionless sleep (b)
for boys and girls at 3 months, 6 months, and 1 year.
interaction between gender and time did not exhibit statistically significant
effects.
Reactivity Univariate Analyses
Figure 3 depicts means and standard errors of REACT-WK (3a), REACTMTR (3b), and REACT-SLP (3c) for boys and girls at W2 and W3. Stability
(from T2 to T3) for reactivity during sleep (REACT-SLP) was r ¼ .21, p ¼ .04.
Reactivity during waking (REACT-WK, REACT-MTR) was unstable. Specifically, stability (from W2 to W3) for REACT-WK was r ¼ .18, ns, and for
REACT-MTR was r ¼ .20, ns. Repeated measures ANOVAs, controlling for
gender as before, revealed that there were no significant developmental
changes in REACT-WK (F[1,71] ¼ 1.84, ns, partial h2 ¼ .03) or in REACT-SLP
(F[1,90] ¼ .44, ns, partial h2 ¼ .01). For REACT-MTR, the repeated measures
effect of time was statistically significant (F[1,71] ¼ 12.16, p ¼ .001, partial
58
REACTIVITY & INFANT SLEEP
FIGURE 3.—Means and standard errors of REACT-WK (a), REACT-MTR (b), and
REACT-SLP (c) for boys and girls at 3 and 6 months. REACT-WK ¼ Total reactivity score during
wakefulness; REACT-MTR ¼ Motor reactivity score; REACT-SLP ¼ Total reactivity score during
sleep.
59
h2 ¼ .15), showing a higher score at W3 than at W2. In all three ANOVAs
(REACT-WK, REACT-MTR, and REACT-SLP), gender and the two-way
interaction between gender and time did not exhibit statistically significant
effects.
Sleep and Reactivity Multivariate Analyses
Next, we explored relations between sensory reactivity and sleep. Using
the entire sample, we calculated Pearson product-moment correlations
between the three reactivity variables at ages 3 and 6 months, and the two sleep
variables at ages 3, 6, and 12 months. Only four of the 36 correlations were
statistically significant: W1 motionless sleep was associated with lower W2
REACT-WK (r ¼ .27, p ¼ .02), and W2 REACT-MTR (r ¼ .25, p ¼ .03),
suggesting that poorer sleep at 3 months was related to higher reactivity at 6
months. Additionally, W1 REACT-MTR was associated with higher W3
motionless sleep (r ¼ .25, p ¼ .03), and W2 REACT-MTR was associated with
higher W3 sleep efficiency (r ¼ .26, p ¼ .03), suggesting that higher reactivity
at 3 and 6 months was related to better sleep at 12 months. These scarce and
conflicting results strengthened our decision to separate the data by gender
and to explore nonlinear (as opposed to linear) relations.
Next, we computed regression equations, separately for girls and boys.
Each regression equation included baseline sleep, reactivity, and squared
reactivity.
Prediction of Sleep as It Develops From 3 to 6 Months-of-Age
In girls, REACT-SLP and REACT-MTR had significant effects on sleep.
Specifically, when predicting motionless sleep by REACT-SLP (R2 ¼ .25,
F[3,35] ¼ 3.90, p ¼ .02, n ¼ 39), significant effects were found for REACT-SLP
(b ¼ .52, p ¼ .02) and REACT-SLP squared (b ¼ .59, p ¼ .01). In this model,
the main effect of sensory reactivity on motionless sleep was positive and the
squared score was negative. This suggests a predominantly positive, concave
downward curve (Aiken & West, 1991), and low reactivity scores predict less
improvement in sleep, although high reactivity scores did not differ from midrange reactivity scores. This model is illustrated in Figure 4a. When predicting
motionless sleep by REACT-MTR (R2 ¼ .32, F[3,33] ¼ 5.25, p ¼ .004, n ¼ 37),
significant effects were baseline sleep (b ¼ .31, p ¼ .04) and REACT-MTR
squared (b ¼ .39, p ¼ .03). In this model, the main effect of reactivity on
motionless sleep was nonsignificant, but the squared score was negative
and significant. This suggests an inverted U-shaped function (Aiken & West,
1991) describes the relationship between sensory reactivity and sleep. That is,
both low scores and high scores in sensory reactivity predict less improvement
in sleep with age. This model is illustrated in Figure 4b. REACT-WK did not
60
REACTIVITY & INFANT SLEEP
FIGURE 4.—Illustrations depicting statistically significant curvilinear effects, controlling
for baseline sleep. (a–d): Effects found in girls, and (e) represents an effect found in boys.
MOT ¼ Motionless Sleep; SEF ¼ Sleep Efficiency; REACT-SLP ¼ Total reactivity score during
sleep; REACT-WK ¼ Total reactivity score during wakefulness; REACT-MTR ¼ Motor reactivity
score.
61
evince statistically significant effects on sleep. In boys aged 3–6 months, there
were no statistically significant effects of sensory reactivity on sleep.
Prediction of Sleep as It Develops From 3 to 12 Months-of-Age
In girls, REACT-MTR and REACT-WK predicted sleep. Specifically, when
predicting motionless sleep by REACT-WK (R2 ¼.32, F[3,30] ¼ 4.61, p ¼ .01,
n ¼ 34), statistically significant effects were those of REACT-WK (b ¼ .47,
p ¼ .01) and REACT-WK squared (b ¼ .39, p ¼ .02), suggesting a predominantly positive, concave downward curve, as explained above. This model is
illustrated in Figure 4c. When predicting motionless sleep by REACT-MTR
(R2 ¼ .36, F[3,30] ¼ 5.54, p ¼ .004, n ¼ 34), statistically significant effects were
baseline motionless sleep (b ¼ .32, p ¼ .04), REACT-MTR (b ¼ .66, p ¼ .001)
and REACT-MTR squared (b ¼ .44, p ¼ .02), again suggesting a predominantly positive, concave downward curve. This model is illustrated in
Figure 4d. REACT-SLP did not evince effects on sleep in girls. In boys, there
were no statistically significant effects of sensory reactivity on sleep.
Prediction of Sleep as It Develops From 6 to 12 Months-of-Age
In girls, there were no statistically significant effects of reactivity on sleep.
In boys, REACT-WK at 6 months predicted sleep efficiency at 12 months
(R2 ¼.26, F[3,36] ¼ 4.24, p ¼ .01). Specifically, REACT-WK (b ¼ .66, p ¼ .002)
and REACT-WK squared (b ¼ .51, p ¼ .01), predicted 12-month sleep
efficiency, again suggesting a predominantly positive, concave downward
curve. This model is illustrated in Figure 4e. Models of REACT-SLP and
REACT-MTR in infants from 6 to 12 months did not reach statistical
significance.
Notably, in most of the regression equations, for all ages, baseline sleep
was not statistically significant. The only exceptions to this were when
predicting motionless sleep in boys from 3 to 6 months (e.g., b ¼ .45, p ¼ .004
when REACT-WK was in the model), and from 6 to 12 months (e.g., b ¼ .52,
p ¼ .001 when REACT-WK was in the model), and in girls from 3 to 12 months
as described above.
DISCUSSION
This is the first study to longitudinally assess sleep and sensory reactivity in
infants during the first year of life using objective sleep and reactivity
measures examined both at the lab during waking and at home during sleep.
We will first address developmental findings and subsequently discuss
relations found between reactivity and sleep.
62
REACTIVITY & INFANT SLEEP
The results reflect the expected maturational trends in sleep–wake patterns
during the first year of life with sleep becoming more consolidated
as manifested by the significant increase in both sleep efficiency and the
percentage of motionless sleep (Burnham, Goodlin-Jones, Gaylor, & Anders,
2002; Tikotzky & Sadeh, 2009). Although sleep efficiency manifested a linear
improvement, motionless sleep exhibited a decline in W3. This suggested that
in our sample, motionless sleep peaked at 6 months, which differs from a
previous report in which motionless sleep exhibited a linear increase during
early development (Burnham et al., 2002). Motionless sleep showed more
overall stability than sleep efficiency. Although motionless sleep has been used
less extensively in the literature, it has been shown to be highly related to
maturation in newborns (i.e., significantly correlated with gestational age and
other anthropometric measures) (Gertner et al., 2002; Sadeh, Dark, & Vohr,
1996), but relatively stable and unaffected by behavioral sleep problems (Sadeh,
1994). Considering our findings, we suggest that perhaps motionless sleep
represents a useful indicator of sleep that is linked to underlying physiological
features related to sensory reactivity in the first year of life. Indeed, motionless
sleep was also better predicted by reactivity, compared to sleep efficiency.
As for reactivity variables, stability reached statistical significance only for
reactivity during sleep (REACT-SLP). A developmental change was shown
only in REACT-MTR. Possibly, gross motor movements significantly increase
in infants from 3 to 6 months, due to maturational processes in motor control
(Thelen, 1995).
Curvilinear relations emerged between reactivity and sleep suggesting
that both hypersensitive and hyposensitive infants are at risk for poorer sleep,
compared to infants with sensory reactivity in the average range. The link
between hypersensitivity and poor sleep has been demonstrated before using
subjective reports of temperament (Carey, 1974; Sadeh et al., 1994). However,
the link between low reactivity and poor sleep has rarely been addressed in the
literature. In our study, this result was replicated across several measures and
assessment waves, especially in girls.
A relevant issue to the current findings is the effect of swaddling on infant
sleep and crying (Franco et al., 2005; Richardson, Walker, & Horne, 2010).
Studies have demonstrated that swaddling improves infant sleep by reducing
spontaneous arousals. The underlying mechanisms explaining the effects of
swaddling on sleep are not clear, but one possibility is that swaddling provides
more consistent sensory stimulation to infants who are low in sensory
reactivity and need a sense of being held. The example of swaddling is only
relevant to touch stimulation and limiting movements (with all their related
sensations), but it may demonstrate a broader principle of a need for constant
stimulation for creating a sense of security.
Another possible explanation is that infants with low reactivity to external
stimuli are less attuned to their surroundings. These children may have a
63
dysregulated sleep schedule and difficulty in consolidating sleep at night
according to environmental cues. Indeed, infants who were more withdrawn
in play interactions with their mothers (i.e., less reactive to stimuli presented
by mothers) were more likely to be characterized by the mothers as generally
unpredictable (Dollberg, Feldman, Keren, & Guedeney, 2006). Additionally,
high thresholds (low reactivity) have been related to adjustment problems
and anxiousness in children (Klein, 1982). Possibly, low reactivity points to
low sensitivity to one’s surroundings, which may bring about problems in
adjustment in sleeping and in waking. Whatever the specific underlying
mechanism, our findings suggest that low reactivity in infancy is related to
poor sleep, and thus should be explored in future studies, along with high
reactivity, as possible negative developmental indicators.
Our findings extend previous literature, largely based on parental
reports, that showed a linear relation between sensory reactivity and sleep
(e.g., Carey, 1974; Ednick et al., 2009; Kelmanson, 2004) with few exceptions
(e.g., Halpern et al., 1994). The present findings are compatible with a view of
temperament as a dimension in which both extremes might interfere with
healthy adjustment (Derryberry & Rothbart, 1997). The curvilinear relation
was not reported in previous studies, and might explain the null results
obtained with objective sleep measures in previous research examining only
linear relations (e.g., Scher, Tirosh, & Lavie, 1998). Our findings only partially
support Carey’s (1974) hypothesis. Although high reactivity was indeed
associated with poor sleep, so was low reactivity. Thus, Carey’s theory of the
role of low sensory threshold (high reactivity) does not fully explain our
findings.
Notably, although some of the curvilinear effects found in this study were
symmetrical, some exhibited a predominantly positive, concave downward
curve (Aiken & West, 1991), suggesting that hyposensitivity was a better predictor
of poor sleep than hypersensitivity. This conclusion is contrary to our
hypotheses and to the linear associations found in previous literature. In
addition, the observed lack of consistent linear correlations among our study
measures is incompatible with previous findings of studies that used subjective
measures for sleep and general temperament. These results might stem partly
from our focus on sensory reactivity, a specific aspect of temperament.
However, studies utilizing subjective measures for reactivity or sensory
thresholds have also shown linear relations with sleep in the past (e.g., Carey,
1974). Possibly, subjective parental reports in previous studies may have
inflated relations focusing on the “hypersensitive” side of the parabola, since
difficult temperament (e.g., hyper-responsiveness) is more easily detected by
parents than reduced or nonresponsiveness (e.g., hypo-responsiveness). The
present findings are important, because even if parents are unaware of
their child’s disturbed sleep, it still has negative consequences (e.g., Scher,
Zukerman, & Epstein, 2005).
64
REACTIVITY & INFANT SLEEP
Our findings also show differences between boys and girls in the relations
between sensory reactivity and sleep. Specifically, most effects emerged for
girls, a result that is compatible with previous findings in a different age group
showing that even if group means are similar, the pattern of associations of
arousal and sleep differ between boys and girls (Fisher & Rinehart, 1990).
Specifically, Fisher and Rinehart showed that baseline sensory activity levels
were related to sleep only in girls. These findings may indicate differential
susceptibility (Belsky & Pluess, 2009) for poor sleep quality between boys and
girls. Gender differences in sleep have been well-documented, particularly
following puberty onset (Manber & Armitage, 1999; Mong et al., 2011). In
comparison to men, women are more likely to have poorer sleep quality and
present more complaints of insomnia. Furthermore, it has been documented
that in young children, sleep is more adversely affected by atopic dermatitis
in girls (Chernyshov, 2012), suggesting that perhaps girls’ sleep is more
vulnerable to skin related sensations. It is possible that our findings reflect
early signs of this differential susceptibility.
It is also important to mention that motionless sleep was predicted only in
girls and sleep efficiency was predicted only in boys. In addition, reactivity
measured at 3 months-of-age predicted sleep only in girls and reactivity at
6 months-of-age predicted sleep only in boys. Again, these results highlight
differences between the sexes even at young ages and underscore the
importance of using multiple measures and assessment waves.
Limitations
Despite a good initial sample size, missing data and the analyses
conducted separately by gender reduced the sample size. Thus, the statistical
power of the study may have been compromised, and thus generalization
may be somewhat tempered. Second, our sample consisted mainly of
educated upper-middle class families, which again limits generalization.
Finally, our sleep assessment was based on nocturnal sleep only, which
excludes understanding of the role of daytime sleep in this context.
Future Directions
Future research on sleep and sensory reactivity in infancy should explore
both extremes of the reactivity continuum using both objective and subjective
measures of these domains. The gender-related differential susceptibility
found in our research should be further explored to see if indeed sleep is
more likely to be affected by sensory stimulation in girls in comparison to
boys. Additional attention should be given to the possibility that sensory
reactivity may play a role in the evolution of sleep problems in infants and
children.
65
CONCLUSION
Notwithstanding the study limitations, this study has important strengths
including: (a) the use of objective assessments for both sensory reactivity and
sleep, thus reducing potential bias caused by parental reports; (b) the
longitudinal design, enabling baseline sensory reactivity to predict change in
sleep over time; and (c) the assessment of both sleep and wake reactivity. This
is the first study, to the best of our knowledge, to assess and to reveal quadratic
associations between sensory reactivity and sleep during the first year of life.
This may explain null or inconsistent results obtained in some previous
studies that explored only linear associations.
ACKNOWLEDGMENTS
We would like to thank all the families who participated and the students
who served as research assistants. Special thanks to Ornit Arbel who
coordinated study logistics.
REFERENCES
Acebo, C., Sadeh, A., Seifer, R., Tzischinsky, O., Hafer, A., & Carskadon, M. A. (2005). Sleep/
wake patterns derived from activity monitoring and maternal report for healthy 1- to
5-year-old children. Sleep, 28, 1568–1577.
Acebo, C., Sadeh, A., Seifer, R., Tzischinsky, O., Wolfson, A. R., Hafer, A., et al. (1999).
Estimating sleep patterns with activity monitoring in children and adolescents: How many
nights are necessary for reliable measures? Sleep, 22, 95–103.
Adams, S. M., Jones, D. R., Esmail, A., & Mitchell, E. A. (2004). What affects the age of first
sleeping through the night? Journal of Paediatrics and Child Health, 40, 96–101. doi:
10.1111/j.1440-1754.2004.00317.x
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury
Park, CA: SAGE.
Allen, K., & Prior, M. (1995). Assessment of the validity of easy and difficult temperament
through observed mother-child behaviours. International Journal of Behavioral Development,
18, 609–630.
Atkinson, E., Vetere, A., & Grayson, K. (1995). Sleep disruption in young children: The
influence of temperament on the sleep patterns of pre-school children. Child: Care, Health
and Development, 21, 233–246.
Belsky, J., & Pluess, M. (2009). Beyond diathesis stress: Differential susceptibility to
environmental influences. Psychological Bulletin, 135(6), 885–908. doi: 10.1037/a0017376
Boismier, J. D., Chappell, P. F., & Meier, G. W. (1974). Wakefulness and REM sleep in human
neonates. Developmental Psychobiology, 7, 304. doi: 10.1002/dev.420070406
Burnham, M. M., Goodlin-Jones, B. L., Gaylor, E. E., & Anders, T. F. (2002). Nighttime sleep–
wake patterns and self-soothing from birth to one year of age: A longitudinal intervention
study. Journal of Child Psychology and Psychiatry, 43, 713–725.
66
REACTIVITY & INFANT SLEEP
Busby, K. A., Mercier, L., & Pivik, R. T. (1994). Ontogenetic variations in auditory arousal
threshold during sleep. Psychophysiology, 31, 182–188.
Carey, W. B. (1974). Night waking and temperament in infancy. The Journal of Pediatrics, 84, 756–
758.
Chernyshov, P. V. (2012). Gender differences in health-related and family quality of life in
young children with atopic dermatitis. International Journal of Dermatology, 51(3), 290–294.
doi: 10.1111/j.1365-4632.2011.04997.x
Chess, S., & Thomas, A. (1996). Temperament: Theory and practice. New York, NY: Bruner-Mazel.
Derryberry, D., & Rothbart, M. K. (1997). Reactive and effortful processes in the organization of
temperament. Development and Psychopathology, 9, 633–652.
Dollberg, D., Feldman, R., Keren, M., & Guedeney, A. (2006). Sustained withdrawal behavior in
clinic-referred and nonreferred infants. Infant Mental Health Journal, 27, 292–309. doi:
10.1002/Imhj.20093
Ednick, M., Cohen, A. P., McPhail, G. L., Beebe, D., Simakajornboon, N., & Amin, R. S. (2009).
A review of the effects of sleep during the first year of life on cognitive, psychomotor, and
temperament development. Sleep, 32, 1449–1458.
Fisher, B. E., & Rinehart, S. (1990). Stress, arousal, psychopathology and temperament: A
multidimensional approach to sleep disturbance in children. Personality and Individual
Differences, 11, 431–438.
Fox, N. A., & Calkins, S. D. (1993). Multiple-measure approaches to the study of infant emotion.
In M. Lewis & J. M. Haviland (Eds.), Handbook of emotions (pp. 167–184). New York, NY:
Guilford Press.
Franco, P., Pardou, A., Hassid, S., Lurquin, P., Groswasser, J., & Kahn, A. (1998). Auditory
arousal thresholds are higher when infants sleep in the prone position. Journal of Pediatrics,
132, 240–243. doi: 10.1016/S0022-3476(98)70438-X
Franco, P., Seret, N., Van Hees, J.N., Scaillet, S., Groswasser, J., & Kahn, A. (2005). Influence of
swaddling on sleep and arousal characteristics of healthy infants. Pediatrics, 115, 1307–
1311. doi: 10.1542/peds.2004-1460
Garcia Coll, C. T., Halpern, L. F., & Vohr, B. R. (1992). Stability and correlates of change of early
temperament in preterm and full-term infants. Infant Behavior and Development, 15, 137–
153.
Gertner, S., Greenbaum, C. W., Sadeh, A., Dolfin, Z., Sirota, L., & Ben-Nun, Y. (2002). Sleep–
wake patterns in preterm infants and 6 month’s home environment: Implications for early
cognitive development. Early Human Development, 68, 93–102. doi: 10.1016/S0378-3782
(02)00018-X
Gjone, H., & Stevenson, J. (1997). A longitudinal twin study of temperament and behavior
problems: Common genetic or environmental influences? Journal of the American Academy
of Child & Adolescent Psychiatry, 36, 1448–1456. doi: 10.1097/00004583-199710000-00028
Gregory, A. M., & Sadeh, A. (2012). Sleep, emotional and behavioral difficulties in children and
adolescents. Sleep Med Rev, 16, 129–136. doi: 10.1016/j.smrv.2011.03.007
Halpern, L. F., Anders, T. F., Garcia Coll, C., & Hua, J. (1994). Infant temperament: Is there a
relation to sleep–wake states and maternal nighttime behavior? Infant Behavior and
Development, 17, 255–263. doi: 10.1016/0163-6383(94)90004-3
Henderson, J. M., France, K.G., Owens, J. L., & Blampied, N. M. (2010). Sleeping through the
night: The consolidation of self-regulated sleep across the first year of life. Pediatrics, 126,
e1081–1087. doi: 10.1542/peds.2010-0976
67
Jimmerson, K. R. (1991). Maternal, environmental, and temperamental characteristics of
toddlers with and toddlers without sleep problems. Journal of Pediatric Health Care, 5, 71–77.
Kagan, J., & Snidman, N. (1991). Infant predictors of inhibited and uninhibited profiles.
Psychological Science, 2, 40–44. doi: 10.1111/j.1467-9280.1991.tb00094.x
Kagan, J., Snidman, N., Zentner, M., & Peterson, E. (1999). Infant temperament and anxious
symptoms in school age children. Development and Psychopathology, 11, 209–224. doi:
10.1017/S0954579499002023
Keener, M. A., Zeanah, C. H., & Anders, T. F. (1988). Infant Temperament, sleep organization,
and nighttime parental interventions. Pediatrics, 81, 762.
Kelmanson, I. A. (2004). Temperament and sleep characteristics in two-month-old infants. Sleep
and Hypnosis, 6, 78–784. doi: 10.1017/S0022112004000515
Kisilevsky, B. S., & Muir, D. W. (1984). Neonatal habituation and dishabituation to tactile
stimulation during sleep. Developmental Psychology, 20(3), 367–373.
Klein, H. A. (1982). The relationship between children’s temperament and adjustment to
kindergarten and Head Start settings. Journal of Psychology: Interdisciplinary and Applied, 112,
259–268.
Korner, A. (1973). Individual differences at birth: Implications for early experience and later
development. In J. C. Westman (Ed.), Individual differences in children. New York, NY: Wiley.
Manber, R., & Armitage, R. (1999). Sex, steroids, and sleep: A review. Sleep, 22(5), 540–555.
Martel, M. M., & Nigg, J. T. (2006). Child ADHD and personality/temperament traits of reactive
and effortful control, resiliency, and emotionality. Journal of Child Psychology and Psychiatry,
47, 1175–1183. doi: 10.1111/j.1469-7610.2006.01629.x
Mezulis, A. H., Hyde, J. S., & Abramson, L. Y. (2006). The developmental origins of cognitive
vulnerability to depression: Temperament, parenting, and negative life events in
childhood as contributors to negative cognitive style. Developmental Psychology, 42, 1012–
1025. doi: 10.1037/0012-1649.42.6.1012
Minard, K. L., Freudigman, K., & Thoman, E. B. (1999). Sleep rhythmicity in infants: Index of
stress or maturation. Behavioural Processes, 47, 189–203. doi: 10.1016/S0376-6357(99)
00065-0
Mong, J. A., Baker, F. C., Mahoney, M. M., Paul, K. N., Schwartz, M. D., Semba, K., et al. (2011).
Sleep, rhythms, and the endocrine brain: Influence of sex and gonadal hormones. The
Journal of Neuroscience, 31(45), 16107–16116. doi: 10.1523/jneurosci.4175-11.2011
Richardson, H. L., Walker, A. M., & Horne, R. S. C. (2010). Influence of swaddling experience
on spontaneous arousal patterns and autonomic control in sleeping infants. The Journal of
Pediatrics, 157, 85–91. doi: 10.1016/j.jpeds.2010.01.005
Rothbart, M. K., & Derryberry, D. (1981). Development of individual differences in
temperament. In M. E. Lamb & A. L. Brown (Eds.), Advances in developmental psychology
(Vol. 1, pp. 37–86). Hillsdale, NJ: Lawrence Erlbaum.
Sadeh, A. (1994). Assessment of intervention for infant night waking: Parental reports
and activity-based home monitoring. Journal of Consulting and Clinical Psychology, 62,
63–68.
Sadeh, A. (1996). Evaluating night wakings in sleep-disturbed infants: A methodological study
of parental reports and actigraphy. Sleep, 19, 757–762.
Sadeh, A. (2004). A brief screening questionnaire for infant sleep problems: Validation and
findings for an Internet sample. Pediatrics, 113(6), E570–E577.
68
REACTIVITY & INFANT SLEEP
Sadeh, A., Acebo, C., Seifer, R., Aytur, S., & Carskadon, M. A. (1995). Activity-based assessment of
sleep–wake patterns during the 1st year of life. Infant Behavior & Development, 18, 329–337.
Sadeh, A., Dark, I., & Vohr, B. R. (1996). Newborns’ sleep–wake patterns: The role of maternal,
delivery and infant factors. Early Human Development, 44, 113–126. doi: 10.1016/
j.jpeds.2010.01.005
Sadeh, A., Hauri, P. J., Kripke, D. F., & Lavie, P. (1995). The role of actigraphy in the evaluation
of sleep disorders. Sleep, 18, 288–302.
Sadeh, A., Lavie, P., & Scher, A. (1994). Sleep and temperament: Maternal perceptions of
temperament of sleep-disturbed toddlers. Early Education and Development, 5, 311–322. doi:
10.1207/s15566935eed0504_6
Sadeh, A., Lavie, P., Scher, A., Tirosh, E., & Epstein, R. (1991). Actigraphic home-monitoring
sleep-disturbed and control infants and young children: A new method for pediatric
assessment of sleep–wake patterns. Pediatrics, 87, 494–499.
Sadeh, A., Mindell, J. A., & Owens, J. (2011). Why care about sleep of infants and their parents?
Sleep Medicine Reviews, 15, 335–337. doi: 10.1016/j.smrv.2011.03.001
Scher, A., & Cohen, D. (2005). Locomotion and nightwaking. Child: Care, Health and
Development, 31, 685–691. doi: 10.1111/j.1365-2214.2005.00557.x
Scher, A., Epstein, R., Sadeh, A., Tirosh, E., & Lavie, P. (1992). Toddlers’ sleep and
temperament: Reporting bias or a valid link? A research note. Journal of Child Psychology and
Psychiatry and Allied Disciplines, 33, 1249–1254.
Scher, A., Tirosh, E., & Lavie, P. (1998). The relationship between sleep and temperament
revisited: Evidence for 12-month-olds: A research note. Journal of Child Psychology and
Psychiatry and Allied Disciplines, 39, 785–788.
Scher, A., Zukerman, S., & Epstein, R. (2005). Persistent night waking and settling difficulties
across the first year: Early precursors of later behavioural problems? Journal of Reproductive
and Infant Psychology, 23, 77–88. doi: 10.1080/02646830512331330929
Shaefer, C. E. (1990). Night waking and temperament in early childhood. Psychological Reports,
67, 192–194.
Spruyt, K., Aitken, R. J., So, K., Charlton, M., Adamson, T. M., & Horne, R. S. C. (2008).
Relationship between sleep/wake patterns, temperament and overall development in
term infants over the first year of life. Early Human Development, 84, 289–296. doi: 10.1016/
j.earlhumdev.2007.07.002
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5 ed.). Boston: Allyn and
Bacon.
Thelen, E. (1995). Motor development: A new synthesis. American Psychologist, 50, 79–95.
Thomas, A., & Chess, S. (1977). Temperament and development. New York, NY: Brunner/Mazel.
Thomas, A., Chess, S., & Korn, S. J. (1982). The reality of difficult temperament. Merrill–Palmer
Quarterly: Journal of Developmental Psychology, 28, 1–20.
Tikotzky, L., & Sadeh, A. (2009). Maternal sleep-related cognitions and infant sleep: A
longitudinal study from pregnancy through the first year. Child Development, 80, 860–874.
doi: 10.1111/j.1467-8624.2009.01302.x
Weissbluth, M. (1981). Sleep duration and infant temperament. The Journal of Pediatrics, 99,
817–819.
Zeanah, C. H., & Fox, N. A. (2004). Temperament and attachment disorders. Journal of Clinical
Child and Adolescent Psychology, 33, 32–41. doi: 10.1207/S15374424jccp3301_4
Zuckerman, B., Stevenson, J., & Bailey, V. (1987). Sleep problems in early childhood—
Continuities, predictive factors, and behavioral correlates. Pediatrics, 80, 664–671.
69